Cyanoaquochromium(III) complexes. Separation and identification of

D. K. Wakefield and. Ward B. Schaap. Inorganic Chemistry catalyzed cleavage of the aquoerythro ion cis- (NHg )5Cr(OH) Cr (NH3 )4OH2s. + + Cl~. A...
0 downloads 0 Views 975KB Size
512 D. K. WAKEFIELD AND WARDB. SCHAAP

Inorganic Chemistry

catalyzed cleavage of the aquoerythro ion

+

~ i s - ( h " ~ ) s C r ( O H ) C r ( N H ~ ) 4+0 H ~Cl~ A

ki e k -1

+

cis-(?iH3)~Cr(OH)Cr(NH~)rCl~ HzO +

B

+ H Z O + J k2

(NHP)&r (OHg)

+

+ cis- (NH3)4Cr(OHg)C12

+

The step which is given the rate constant K z would probably be the composite of a slow step and one or more rapid steps, including the neutralization of the OH- which served as the bridge. If the intermediate cis-chloroerythro ion is treated as a reactive intermediate, the general rate law for the reaction by this path is

Equation 7 is equivalent to the chloride-dependent term of eq 6, where k' = k l k z / ( k - , k2). It seems useless to speculate on the relative magnitudes of the rate constants of the proposed mechanism at this time, since we expect to be able to isolate the cis-chloroerythro ion and measure its reaction rates directly. Absorption maxima and minima for the various monomeric complexes synthesized in the course of this study are given in Table IV.

+

Acknowledgment.-The authors are grateful to the Sational Science Foundation for financial support of this research (Grant GP-5425).

CONTRIBUTION FROM THE DEPARTMENT OF CHEMISTRY, IXDIAXA UNIVERSITY, BLOOMIXGTON, INDIANA47401

Cyanoaquochromium(II1) Complexes. Separation and Identification of the Neutral and Cationic Cyanoaquo Complexes of Chromium(II1) and Aquation Kinetics of the Monocyanopentaaquochromium(II1) Ion' BY D . K . WAKEFIELD2

AND

WARD B. SCHAAP

Received August 16, 1968 The neutral and cationic cyanoaquochromium(II1) complexes formed as intermediates in the acid-assisted aquation of hexacyanochromate(II1) ion were separated using ion-exchange techniques. Species isolated include the uncharged 1,2,3tricyanotriaquochromium(II1) complex and the cis-dicyanotetraaquo- and cyanopentaaquochromium(II1) cations. The aquation kinetics of the cyanopentaaquochromium(II1) ion were studied in acidic aqueous solutions and were found to proceed via an acid-independent pathway and also an acid-catalyzed pathway involving an equilibrium protonation of the k ~ K ~ , [ H + ] ) C t l ( l Ke,[HL]), where complex. The rate of the aquation is described by the rate law -dCt/dt = (ko Ct is the total concentration of the cyanopentaaquochromium(II1) ion in both protonated and nonprotonated forms. At 25" in perchloric acid-sodium perchlorate solutions of ionic strength 2.0 M , the rate parameters are ko = (1.1Z!C 0.1)X sec-l for the acid-independent path, k E = (3.2 = 0.1) X 10-3 sec-1 for the aquation of the protonated species, K,, = 0.185 -I: 0.006 1. mol-' (for the protonation equilibrium), and ~ H K , = , (5.90 +C 0.05) X sec-l 1. mol-'. The aquation was also studied a t 15, 35, and 45' and the activation parameters were calculated. For the acid-independent (k,) pathway, AHo* = 26.9 +C 0.3 kcal mol-' and A&* = 8.9 +C 1.1eu. For the acid-assisted ( ~ E K , , pathway, ) AH* = 20.2 f: 0.1 kcal mol-' and AS* = -5.5 i: 0.3 eu. The pseudo-first-order rate constants calculated from these activation parameters agreed with the experimental values with an average deviation of 1.37,.

+

Introduction In acidic aqueous solutions, the hexacyanochromate(111) ion aquates via a series of stereospecific reactions t o give, finally, the hexaaquochromium(II1) cation.1ba2b3 During the aquation the cyanide ligands are replaced by water molecules in a stepwise manner, forming a pre(1) (a) This work was supported by the U . s. Atomic Energy Commission under Contract KO. A T ( l l - 1 ) - 2 5 6 (Document No. COO-256.84). (b) Presented in part before t h e Division of Inorganic Chemistry a t t h e 150th National Meeting of t h e American Chemical Society, Atlantic City, N. J., Sept 1965; a t t h e 9th International Conference on Coordination Chemistry, St. Moritz-Bad, Switzerland, Sept 1966; and a t t h e l53rd National Meeting of t h e American Chemical Society, Miami Beach, Fla., April 1967. (2) Taken in part from a thesis submitted by U . K. Wakefield t o the Graduate School of Indiana University in partial fulfillment of t h e requirements for t h e Ph.D. degree i n chemistry, 1967. (3) (a) R. Krishnamurthy, W.B. Schaap, and J. R . Perumareddi, Inoug. Chem., 6, 1338 (1967); (b) R. Krishnamurthy, Thesis, Indiana University, 1966.

+

viously unknown4 series of intermediate cyanoaquochromium(II1) complexes having the general formula Cr(CN).(H20)6--n3-n. Under the proper hydrolysis conditions each of these aquation intermediates can be isolated in solution by ion-exchange techniques. This paper decribes the separation, identification, and spectral properties of the neutral and cationic intermediates and the aquation kinetics of one of them, the monocyanopentaaquochromium(II1) ion. The aquation of monocyanochromium(II1) ion (eq 1) Cr(HgO)jCN*+

+ H + + HgO

--f

+ HCN

Cr(H20)s3+

(1)

(4) (a) T h e cyanopentaaquochromium(Il1) complex was prepared by a different method and previously reported by J. H. Espenson and J , P. Birk, J . A m . Chem. SOL.,87, 3280 (1965). (b) A method of preparing cyanopentaaquochromium(II1) ion a t higher concentrations is given by J. P. Birk and J. H. Espenson, ibid., 90, 2266 (1968).

CYANOAQUOCHROMIUM(III) COMPLEXES513

VoL 8 , No. 3, March 1969 was studied over a range of hydrogen ion concentrations and a t ionic strengths of 1.5 and 2.0. After completion of this work,lb Birk and Espenson6 reported the results of their study of the aquation of monocyanochromium(111) ion which employed somewhat different conditions of acidity and ionic strength than used in the present study. Monosubstituted pentaaquochromium(II1) complexes form an interesting series of compounds in which basic ligands such as azide,6 fluoride,7 hypophosphite,8 sulfate, and acetatelo show an acid-catalyzed aquation pathway which has a first-order dependence on hydrogen ion concentration, presumably due to protonation of the ligand. Complexes of these ligands, with the exception of hypophosphite, also show an acid-independent pathway which may involve intramolecular proton transfer in the case of more basic ligands such as azide and fluoride. The acid-catalyzed pathway is not present in the case of complexes with less basic anions.11-16 The monocyano complex is of particular interest because the high basicity of cyanide ion makes it probable that protonated intermediates would be involved in both the acid-catalyzed and acid-independent mechanisms and that their presence might be detected directly as well as inferred from their effects on the parameters of the reaction pathways.

Experimental Section Reagents.-Deionized water and reagent grade chemicals were used in all procedures. Stock solutions of sodium perchlorate used t o adjust the ionic strength were prepared from sodium carbonate and perchloric acid. Potassium hexacyanochromate(111) was synthesized and purified by the method given by Bigelow." Analytical grade anion-exchange resin (Bio-Rad, AG 1-X4, 100-200 mesh), converted to the perchlorate form by washing with 1 .O M sodium perchlorate solution, and analytical grade cation-exchange resin (Bio-Rad, AG 50W-X4, 100-200 mesh) in the sodium form were used in the separation procedures. The pH of the cation-exchange resin was adjusted to about 2.2 by equilibrating the resin in a beaker with a little 0.01 M perchloric acid solution, which served t o convert a small fraction of the resin to the hydrogen cycle. The acidified sodium-form resin was washed with deionized water and its pH was tested by immersing a pH electrode into the wet resin. Hydrolysis Mixture.-A solution containing the neutral and cationic cyanoaquochromium(II1) complexes was prepared by combining potassium hexacyanochromate with perchloric acid in a molar ratio of 1:5. I n a typical experiment, 0.65 g (2.00 mmol) of potassium hexacyanochromate was dissolved in 250 ml of 0.04 M perchloric acid. The hydrolysis solution was kept in the dark a t room temperature until the pH of the solution rose to 2.2, which took approximately 44 hr. A stream of nitrogen was continuously bubbled through the solution to sweep out the

liberated molecular, hydrogen cyanide. The deep red solution was cooled to 0' and the precipitated potassium perchlorate was removed by filtration. Separation and Identification of Species.-The chilled hydrolysis mixture was first passed through an anion-exchange column to remove any possible anionic cyanoaquochromium(II1) complexes. No colored anionic complexes were detected on the resin, indicating their absence from the mixture. The various neutral and cationic complexes in the effluent from the anion-exchange column were loaded onto a cation-exchange column with resin-bed dimensions of about 1.5 X 20 cm. Both columns were jacketed and cooled below 1.0' by means of circulating ice water and both were shielded from the light wtih aluminum foil. As the cationic species were adsorbed onto the cation-exchange column, brightly colored bands were formed a t the top of the column. A deep orange uncharged species was not adsorbed and moved through the column with the solution. The lowest adsorbed band a t the top of the column was peach colored, above this was a small yellow one, next a pink-violet band, and then a purple band a t the top. These bands could be separated and eluted individually from the column by gradually increasing the concentration of sodium perchlorate used as the eluting agent. The effluent from the cation column was collected in 25- or 50-ml fractions, the pH was measured, and the fractions were stored in the dark a t 0" for spectral and chemical analysis. Each fraction collected was analyzed for chromium(111)ion concentration in order to obtain the elution curve for the separation shown in Figure 1. The separation and identification of the individual species are described below. (1 ) Tricyanotriaquochromium(III),Cr (CN)3(H20)3.-This uncharged complex is not retained by either an anion or a cationexchange resin and emerges from the cation-exchange column as a deep-orange solution, the last traces of which can be washed off the column with 0.05 M sodium perchlorate adjusted to p H 3. The composition of this species was confirmed by chromium and cyanide analyses (see below) which consistently gave chromium to cyanide ratios in the range 1:3.11 t o 1:2.92. I n the visible and ultraviolet regions the absorption spectrum of tricyanotriaquochromium(II1) is independent of p H in the range 3.0-4.5 and has absorption maxima (in mp) and corresponding molar absorbancy indices (in 1. mol-') as follows: 687 (0.2), 620 (0.3), 467 (115), 360 ( 2 5 . 2 ) , 236 (2.1 X lo3); and 227 (2.1 X lo3). The tricyano complex is extremely soluble in water but usually forms insoluble orange solids, thought to be polymers, when solutions are concentrated by evaporation, by freeze-drying techniques, or even by repeated freezing and thawing. T h e optimum pH range for storing the tricyano complex appears to be 3.5-4.5. The tricyano complex was found to constitute apout 38y0 of the hydrolysis mixture prepared as described above.

5

kP3" IO? M 3

(5) J. P . Birk and J. H. Espenson, Inovp. Chem., 7 , 991 (1968). (6) T. W. Swaddle and E. L. King, ibzd., 3, 234 (1964). (7) T. W. Swaddle and E . L. King, i b i d . , 4, 532 (1965). (8) J. H. Espenson and D. E. Binau, ibid., I,1305 (1966). (9) J. Finholt and S. N. Deming, ibid., 6, 1533 (1967). (10) E. Deutsch and H. Taube, ibid., 1 , 1532 (1968). (11) C. Postmus and E . L. King, J. Phys. Chem., 69, 1216 (1955). (12) F. H. Guthrie and E. L. King, Inovg. Chem., 3, 916 (1964). (13) J. H. Espenson, ibid., 3, 968 (1964). (14) M. Ardon, ibid., 4, 372 (1965). (15) K. Jones and J. Bjerrum, Acla Chem. Scand., 19, 974 (1965). (16) T. W. Swaddle, J . Am. Chem. SOL.,89, 4338 (1967). (17) J. H. Bigelow, Inoyg. Syn., 2, 203 (1946).

I

200

400 Ebole

600 Volume.

800

ml

Figure 1.-Elution curve for the ion-exchange separation of neutral and cationic cyanoaquochromium(II1) complexes: I, C ~ ( C N ) S ( H Z O11, ) ~ ;Cr(CN)2(Hz0)4+;111,unidentified species; IV, Cr(CN)(Hz0)E2+;V, Cr(H20)s3+.

514 D. K. WAKEFIELD AND WARDB. SCHAAP (2) Dicyanotetraaquochromium(II1) ion,,Cr(CN)z(HzO)c+.The peach-colored band containing this species was eluted from the cation-exchange column with 0.10 M sodium perchlorate solution a t p H 2.5. Analysis of these solutions gave chromium t o cyanide ratios of 1: 1.99, 1: 2.03, and 1: 1.96 in three trials. T h e visible and ultraviolet absorption spectrum of the dicyano complex is independent of p H in the region from 2.0 to 4.0. The position of the maxima (in m p ) and corresponding molar absorbancy indices (in 1. mol-' cm-l) in dilute sodium perchlorate are 535 (30.2) sh, 474 (45.3), 464 (45.5), and 378 (25.8), with strong absorption in the ultraviolet region. On freeze-drying a solution of the dicyano complex, a powder was obtained which was readily soluble in water, methanol, and ethanol. T h e optimum p H for storing this complex appears t o be about 3 . For the hydrolysis procedure described above, about 307, of the product was in the form of the dicyanotetraaquochromium(II1) ion. (3) Unknown Species.-When t h e eluent concentration was increased in successive steps from 0.1 t o 0.4 2i' sodium perchlorate, a small concentration of an unknown species was eluted from the column which contained 0.67, of t h e total chromium present in the hydrolysis mixture. Separations carried out a t various initial concentrations of potassium hexacyanochromate ranging from t o lo-' M showed t h a t the amount of this species formed depends greatly on the initial concentration of hexacyanochromate used, which may be taken as evidence of its polymeric nature. T h e species is not produced in detectable amounts if very dilute solutions and long reaction times are used. The ultraviolet and visible spectral characteristics of the species are intermediate between those of the dicyano and tricyano complexes, though the cyanide t o chromium ratio appears t o be 2 . 0 : l . Because i t is eluted just before the cyanopentaaquo complex, we believe t h a t the species probably has an over-all charge of + 2 and t h a t it is most likely a dimer. Application of previously discussed spectral criteria for distinguishing between cis and trans isomers of chromium complexes3 clearly indicates t h a t this yellow species is not the Irans-dicyanotetraaquochromium( I11) ion. (4) Cyanopentaaquochromium(II1) Ion, Cr(CN)(HzO):z+.This complex was found t o constitute 207, of the hydrolysis mixture and could be eluted from the column with 0.6 &f sodium perchlorate solution adjusted to p H 2.0. Chromium t o cyanide ratios of 1:0.992, 1:1.103, and 1:1.009 were determined in separate experiments for this pink, permanganate-colored species. The visible and ultraviolet spectra of cyanopentaaquochromium(111) ion are independent of p H in t h e regions from 2 to 3.5. The positions of the maxima (in mp) and corresponding molar absorbancy indices (in 1. mol-' cm-l) in 0.6 iZf sodium perchlorate are: 800 (0.30), 525 (26.0), and 393 (20.5). T h e strong absorption band in the ultraviolet shows a point of inflection (a, = 268 1. mol-' cm-l) a t 235 mp. The positons of the absorption maxima are identical with those reported by E ~ p e n s o n . ~ Dilute solutions (pH 2 ) of the monocyano complex can be stored a t 0" for several weeks without serious decomposition. ( 5 ) Hexaaquochromium(II1) Ion, C r ( H ~ o ) ~ ~ + . - T hhydrated e chromium(II1) cation was washed off the column with a solution containing 0.5 M perchloric acid and 2.5 Af sodium perchlorate. It was identified by its typical three-band ultraviolet and visible absorption spectrum and comprised about lly0 of the hydrolysis mixture. were left on t h e column. Traces of hydrolytic The total amount of chromium(II1) ion eluted from the cation column amounted to more than 98T0 of t h a t originally loaded onto the column. Kinetic Measurements.-The aquation of monocyanochromium(II1) ion was followed spectophotometrically using a Cary Model 14 recording spectrophotometer for all measurements. Reactions were carried out directly in a 10-cm waterjacketed silica spectrophotometer cell. As a usual procedure an (18) M. Ardon and R . A. Plane, J . A m . Chem. Soc., 81, 3197 (1959). (19) J. S. Laswick and R. A. Plane, ibid., 81, 3564 (1959).

Inorganic Chemistry aliquot of the stock solution of the monocyanochromium(II1) ion and a solution containing perchloric acid and enough sodium perchlorate t o achieve the desired ionic strength were mixed in a darkened constant-temperature water bath and then transferred to the spectrophotometric cell with a fast draining pipet. In all experiments the hydrogen ion concentration was at least 50 times the concentration of the complex, which was kept low and in the M. range 5 X lo-*t o 3 X Reactions were followed a t either 525 or 240 m p . T h e molar absorbancy indices of monocyanochromium(II1) ion and of hexaaquochromium(II1) ion a t 525 m p are 25.9 and 7.34 1. mol-' cm-l, respectively. At 240 m p the molar absorbancy index of the monocyanochromium(II1) complex changes with acidity and ionic strength and is approximately 182 1. mol-' cm-1. T h e corresponding index for hexaaquochromium(II1) ion is 5.1 1. mol-' cm-1. T h e rate of aquation was found to be independent of the length of time the aquation solutions were exposed to monochromatic light a t 525 mp. When reactions were followed a t 240 mp the wavelength and slit settings were not changed during the experiment because in this region of the spectrum the absorbance changes sharply with wavelength. Reaction rates determined at these two wavelengths were in good agreement. In all experiments the reactions went t o completion and gave, after 8 or more half-lives, a final spectrum identical with t h a t of the hexaaquochromium(II1) ion. All reactions were followed for 3 or 4 half-lives and gave good pseudo-first-order rate plots of log ( A t - A , ) 21s. time, where A t is the absorbance a t time t and A , is the absorbance after more than 8 half-lives. A computer programz0 which was adapted t o the available C D C 3600 computer was used to calculate the least-squares best slopes for the straight-line plots of In ( A t - A,) v s . time and has used t o obtain the pseudo-first-order rate constants. T h e rate constants were calculated for the first, for the first and second, and for t h e first, second, and third half-lives. All three calculations were in good agreement and the rate constant reported for each experiment is the average of the three values obtained. Chemical Analysis.-The aquated solutions were analyzed for the hydrogen ion concentration, for the total ionic strength, and periodically for the chromium(II1) ion concentration. The concentrations of the chromium complexes in solution were determined by analyzing t h e solutions for chromium spectrophotometrically following oxidation of the chromium(II1) to chromate ion with hot alkaline hydrogen peroxide (a, = 4.83 X lo3 1. mol-' cm-' for C r 0 2 - a t 372 mp).21 T h e amount of bound cyanide in the cyanoaquochromium(II1) complexes was determined by the modified Liebig titration method. In order t o do this, the compounds were first decomposed in 0.8 A4 sodium hydroxide and boiled t o precipitate chromic hydroxide, which was subsequently removed by filtration prior to t h e titration of the filtrate for cyanide ion. T h e ionic strength was determined from the total electrolyte concentration using standard ion-exchange techniques involving a cation-exchange resin in the hydrogen form. Concentration values are given at room temperature in all cases.

Results Separation.-Under the hydrolysis conditions described in the previous section, a t least five different chromium(II1)-containing species exist in the final solution. These species can be cleanly separated by means of a cation-exchange resin, as shown by the elution curve given in Figure 1. Successive fractions of each peak of the elution curve, including leading and trailing fractions, exhibited visible and ultraviolet absorption spectra with identical molar absorbancy indices (20) K. B. Wiberg, "Physical Organic Chemistry," John Wiley Inc., New York, N. Y., 1964, p 554. (21) G . U '. Haupt, J. Res. A'ali. But'. Sid., 48, 414 (1952).

8;

Sons,

Vol. 8, No. 3, March 1969

CYANOAQTJOCHROMIUM(III) COMPLEXES515

over the wavelength scanned. For example, successive fractions of the first peak, corresponding to the tricyanochromium(II1) complex, gave molar indices of 115.6, 115.6, 114.8, and 114.9 a t a wavelength of 467 mp. Successive fractions of the second peak, corresponding to removal of the dicyanochromium(II1) complex, gave indices of 45.6, 45.6, 45.8, and 45.3 a t 464 mp. This reproducibility indicates that only one species is included in each peak of the elution curve and, more specifically, that only one of the two possible geometric isomers of Cr(CN)3(H20)3 and of Cr(CN)2(H20)4+ are formed. On the basis of their spectra and the reaction rate data obtained for each step in the acid-catalyzed aquation of hexacyanochromate(II1) ion, these complexes have been assigned a cis configuration. The arguments for these assignments are presented in detail elsewhere.3 Spectra.-The visible and ultraviolet absorption spectra for 1,2,3-Cr(CN)3(H20)3 or ci~-Cr(CN)3(H20)3, cis-Cr(CN)z(H20)4+, and Cr(CN) (H20)52+ are shown in Figure 2. The spectrum of Cr(Hz0)63+is also included for the sake of comparison. As cyanide ligands are successively replaced by water molecules, the frequencies of both absorption bands attributed t o d-d transitions shift markedly toward lower energies. The observable absorption spectrum of ll2,3-Cr(CN)3(H20)3 consists of two relatively narrow ligand field bands which are symmetrical without apparent broadening or splitting, in agreement with the ligand field predictions for a pseudo-cubic, trigonal (C3") field.3 The sums of the dipole moments of the ligands are identical along

each of the Cartesian axes in a 1,2,3-trisubstituted complex and the five d orbitals maintain the same relative energies that they possess in fields of cubic (Oh) symmetry. Because the ligand field strengths of cyanide and water differ greatly, the other mixed cyanoaquochromium(II1) complexes exhibit noticeable splitting or broadening of the first (lowest energy) spin-allowed d-d band.3s22 I n the case of cis-Cr(CN)z(H20)4+,the first band shows a well-resolved shoulder, and i t is the only cis-chromium(II1) complex known which exhibits such splitting. In the case of Cr(CN) (Hz0)b2+, on the other hand, the first band shows no splitting but is instead broadened and unsymmetrical. The explanations and implications of these observations are presented elsewhere,3 together with a discussion of criteria for predicting the spectra of isomers of various chroniium(II1) complexes. Aquation of Cyanopentaaquochromium (111) 1on.The aquation of monocyanochromium (111) ion was studied over a range of acidities from 0.05 to 1.6 perchloric acid at an ionic strength of 2.0 adjusted with sodium perchlorate. Repetitive scans of the absorption spectrum of the aquating solution showed the presence of three isosbestic points in the visible region of the spectrum which remain throughout the reaction. In a solution containing 1.0 M perchloric acid and 1.0 A4 sodium perchlorate, the isosbestic points are located a t 577, 447, and 417 mp. The positions of the points near 577 and 417 mp, however, vary slightly with acidity, indicating a slight acid dependence of the absorption spectrum of the monocyanochromium(II1) ion. The pseudo-first-order rate constants for the aquation reaction a t 15, 25, 35, and 40" were determined as described in the Experimental Section and are presented in Table I. Although the aquation is acid catalyzed, the rate of the reaction exhibits less than a simple first-order dependence upon the hydrogen ion concentration, as shown in Figure 3. I n the concentration range studied, the hydrogen ion dependence of the pseudo-first-order constants, k o b s d , indicates the presence of two reaction pathways for the aquation. The positive extrapolated value of k o b s d a t zero hydrogen ion concentration (Figure 3) indicates the presence of a reaction pathway whose rate law does not involve hydrogen ion from the solution and whose stoichiometry can be represented by ko

+

Cr(H20)6CNZ+ 2H20 --f Cr(H20)$+ f H C N

+ OH-

(2)

The rate of the other pathway depends on the hydrogen concentration in a limiting way which is consistent with a two-step mechanism involving a rapid and reversible equilibrium protonation followed by a rate-limiting aquation of the protonated species

+ H+

Cr(H20)&N2+ Wove Length in

mv

Figure 2.-Visible and near-ultraviolet spectra of chromium(111) complexes a t 25': A, 1,2,3-Cr(CN)3(Hz0)3; B, cisCr(CN)s(H20)4+; C, Cr(CN)(Hn0)52+;D, Cr(H20)e3+.

+

Kea fast

Cr(H20)6CNH3+

(3)

kH

C r ( H 2 0 ) 6 C N H a + H 2 0 +C T ( H ~ O )f ~ ~H+C N

(4)

(22) (a) J. R. Perumareddi, J. Phys. Chem., 71, 3144 (1967); (b) J. R. Perumareddi, ibid., 71, 3155 (1967).

516 D. K. W A K E F I E L D

VALUES O F

AND W A R D

THE

.M 0.990 1.18 1.40

1.56 1.83

Inorganic Chemistry

TABLE I PSEUDO-FIRST-ORDER RATECONSTANT MEASURED AT 2 5 . 0 ° . I = 2.06 M a

16.00, I = 2 . 0 M =

W+1,

B.SCHAAP

W+l,

104kabed,

1.58 1.74 2.00 2.26 2.44

1O4kobsd,

-l!

sec-1

0.106 0.208 0.413 0.591 0.609 0,813 0.994 1.21 1.22 1.41 1.44 1.61 1.61 1.70

0.717 1.30 2.34 3.26 3.39 4.24 4.97 5.89 6.07 6.73 6.86 7.40 7.28 7.79

sec-'

A S I O N I C STRENGTH OF

35.0°, I = 2.01 .'do. 1O4kobsd, [H+I, M sec-1

0.109 0.210 0,407 0.596 0.811 1.00 1.01 1.21* 1.21

2.41 4.10 7.39 10.2 13.2 15.4 15.6 18.5 17.6

2.0

40.0°. I = 2 . 0 2 M a W+l, 104kubsd A4 sec-1

0.0521b 0.108 0.112b 0.209 0.316 0.410 0.523b 0.607 0.806 0.993 1,200b

2.65 4.40 4.64 7.29 10.2 13.3 15.0 17.8 22.0 26.4 30.9

Summaryc Constants

25.0'

35.00

40.0'

lo%, sec-l 1.1 i 0.1 4.0 f0.4 10.8 f0 . 8 sec-' 3.2 f0 . 1 7 . 5 f0 . 5 13 f 2 . 0 0 . 2 5 f 0.02 K,,, 1. mol-' 0.186f0.006 0.23 zk 0.05 5 . 9 0 i 0.05 18.9 f 0 . 0 3 31.7 zk 0 . 8 lO4k=Ke,, sec-' 1. mol-' a The ionic strength was maintained at 2.0 with sodium perchlorate. [Cr(H20)5CN2-]o= 6 X M. In all other experiments, The confidence interval is determined from the appropriate variance calculation for the initial concentration was 2 6 X M, the slope and intercept of the reciprocal plot at 1.0 unit of standard deviation.

Under pseudo-first-order conditions of constant hydrogen ion concentration, combination of eq 2-4 leads to the rate law

where Ct represents the total concentration of the monocyanochromium(II1) complex in both the protonated and nonprotonated forms, ko and k H are the first-order rate constants for reactions 2 and 4, respectively, and K e , is the equilibrium constant for the protonation reaction (eq 3 ) . Equation 5 can be rearranged to give 1 1 ____ - ___ kobsd - ko kH - ko

+

H'.

(kH

1 1 - ko)Keq [H']

(6)

M

Figure 3.-Dependence on hydrogen ion concentration of the pseudo-first-order rate constant for the aquation of Cr(HzO)BCNZ+a t 28.0' and an ionic strength of 2.06 M.

A plot of l/(kobsd - ko) us. l / [ H + ] ,which should be linear, can be used to evaluate the individual rate parameters, ko, k H , and Keq. If ko is known, K,, can be evaluated from the ratio of the intercept to the slope of the plot, and k~ can be calculated from the intercept. An iterative calculation method was used to obtain the rate parameters. -4 computer program was developed which calculates the least-squares best line for a plot of l / ( k o b s d - ko) 11s. l/[H+] and also varies ko to find the straight line which has the minimum standard deviation. Trial values of ko were obtained from plots of kobsd vs. [H+](Figure 3). The relative per cent errors in the pseudo-first-order rate constants were judged to be constant and the data points were weighted according to the squares of the corresponding hydrogen ion concentrations. Plots of l / ( k o b s d - ko) vs. l/[H+] using the best value of ko obtained from the iterative calculations were excellent straight lines. The values of koj k ~ Keq, , and kHKeq calculated for the aquation of monocyanochromium(111) ion a t 25, 35, and 40" using all data points a t each temperature are summarized in Table I. The activation parameters corresponding to the individual rate coefficients ko and k H and to the product of k&,, were determined using the Eyring transitionstate equation ( K = 1). The pseudo-first-order rate constants calculated from the activation parameters agree well with the experimental values, the average deviation being only 1.3%. Effect of Ionic Strength on the Aquation Rate.-The rate of aquation of monocyanochromium(II1) ion at a given hydrogen ion concentration was found to increase significantly with increasing ionic strength. The pseudofirst-order rate constant for the aquation reaction was measured at various concentrations of hydrogen ion a t

Vol, 8, No. 3, March 1969

COMPLEXES 517

cYANOAQUOCHROMIUM(III)

25" and a t an ionic strength of 1.5. The data are presented in Table 11. Although the actual ionic strengths of the aquating solutions (determined by analysis) deviated slightly from l .50, the pseudo-first-order rate constants (kobsd) were corrected to an ionic strength of TABLE I1 VALUESOF THE PSEUDO-FIRST-ORDER RATECONSTANT MEASURED AT A VARIETY O F IONIC STRENGTHS" FOR = 2.6 X M [Cr(HzO)~CN2+]o IO

[H+I, M

1.55 0.412 (1.50) 1.55 0.611 (1. 5 0 ) 1.52 0.010 (1.50) 1.00 1.50 1.48 0.106 (1.50) 1.46 0.973 (1.50) All measurements ionic strength of 1.50. perchlorate. Q

[H*I,

104kabad,

sec-1

IO

M

104k0bsd, sec -1

1.95 2.67 0.212 1.74 (1.90)b 2.60 0.112 0.919 2.73 2.60 0.411 2.87 (2.68)b 2.51 0.614 3.94 4.17 2.44 0.818 4.95 (4.17)b 2.37 1.03 5.72 4.20 2.32 1.02 5.56 0,562 2.21 1.42 6.83 (0.568)b 0.961 0.381 1.22 3.88 (3.93)b made a t 25'. Value interpolated to an c Ionic strength maintained with sodium

1.50 by linear interpolation between the experimental values of kobsd and the corresponding values of kobsd calculated from the rate parameters determined a t an ionic strength of 2.06 (Table I). The rate parameters ko, k H , and K,, calculated for the aquation reaction a t 25" were 9.2 X 10-B sec-l, 3.2 X l o m 3sec-l, and 0.14 1. mol-', respectively, a t I = 1.50, and were 1.1 X sec-', 3.2 X sec-l, and 0.185 1. mol-', respectively, a t 1 = 2.06. Whereas the first-order rate constants ko and k H appear to be constant within experimental error in this range of ionic strength, the concentration equilibrium constant for the protonation reaction appears to vary, presumably owing to changes in the activity coefficient quotient with changes in ionic strength in the expression K G = K a ( Y C r C N * '1 ( YH +)/(YCrCNH*+). The aquation reaction was also studied a t 25" a t a variety of other ionic strengths ranging from 0.9 to 2.7. The pseudo-first-order rate constants obtained are included in Table 11. If the rate coefficients ko and k~ are assumed to be constant over this range of ionic strengths, as the data indicate, the equilibrium constant can be calculated by solving eq 5 for K,,. The values of K,, obtained in this manner describe a fairly smooth curve when plotted w. ionic strength (Figure 4). (A similar increase in the value of K,, for the protonation of the acetatopentaaquochromium(II1) ion with increased ionic strength was observed by Deutsch and Taubeloin lithium perchlorate solutions.)

Discussion In the acidity range studied, the aquation of monocyanochromium (111) ion proceeds oia two pathways, 23 (23) A third pathway with inverse hydrogen ion dependence has been detected by Espenson,6 who carried out studies a t lower hydrogen ion concentrations. This pathway is negligible in t h e acidity range covered in this study.

0.20

-

i"

t

f 0.10-

0.0

I

I

I

I

I

0,5

ID

15

2.0

2.5

Ionic Strength

Figure 4.-The equilibrium constant for the protonation reacCr(H20)sCNH3+as a function of tion Cr(H20)&NZf 4- H + ionic strength a t 25.0".

an acid-catalyzed and an acid-independent pathway. The hydrogen ion dependence of the acid-catalyzed pathway is less than first order, as seen in Figure 3. This lack of linearity is interpreted in terms of an equilibrium protonation of the complex followed by the aquation of the protonated complex, reactions 3 and 4. It has also been suggested5 that such a lack of first-order dependence might be due t o medium effects arising from variations in activity coefficients as sodium perchlorate is replaced by perchloric acid in solutions of constant ionic strength. This possibility is discussed more fully below. In a preequilibrium protonation mechanism, a significant portion of the monocyanochromium(II1) complex might be present in the protonated form a t higher acid concentrations. Since the spectrum of the protonated complex should differ from that of the unprotonated complex, spectral changes should occur as the hydrogen ion concentration is increased and were in fact observed. Variations in activity coefficients alone should not cause such spectral changes in the absence of protonation. In the case of the monocyano complex, unfortunately, the spectral changes upon addition of acid, though definite, were too small to allow a precise value to be measured for the protonation equilibrium constant. In other experiments, changes in spectra with variations in acidity were found to be more pronounced for the dicyanotetraaquochromium(II1) and the tricyanotriaquochromium(II1) complexes, where proton association is expected to be greater because of the decreased positive charge. For these latter two complexes, the protonation equilibrium constants were determined from the changes in their spectra with hydrogen ion concentration2 and were found to be 0.55 f 0.22 1. mol-' for the dicyanochromium(II1)

AND WARDB. SCHAAP 518 D. K. WAKEFIELD

Inorganic Chemistry

TABLE 111 RATEA N D ACTIVATIOKPARAMETERS FOR THE AQUATION OF Cr(HzO)sCh'2+ Ionic strength, medium

1 . 0 M, LiC104"

a

R a t e parameters a t 2 5 O

AH*, kcal mol-'

1.3 X sec-l 27.9 f 1 . 4 kl = 5.2 X 1. mol-' sec-l 17.05f0.73 1.0 M , SaC10p ko = 1 . 6 X 10-5 sec-' 25.6 f 1 . 0 kl = 3.1 X 1. mol-' sec-l 19.76 f 0.48 2 . 0 M , NaC104 ko = 1.1 X 10-5 sec-I 26.9 i 0 . 3 kHK,, = kl = 5 , 9 X lo-' 2 0 . 2 i0 . 1 1. mol-' sec-l kH = 3 . 2 X 10-8 sec-l 16.2 f2.0 T h e rate parameters are extrapolated to 25' from higher temperatures.

ko

=

ion and 1.1 i 0.4 1. mol-' for the tricyano complex a t 25" and an ionic strength of 2.0 (sodium perchlorate). The corresponding value of 0.19 1. mol-' for the monocyanochromium(II1) ion, determined from the kinetic data, is reasonable in comparison with the values for the higher cyanide complexes. The aquation of monocyanochromium(II1) ion has also been studied by Birk and Espenson5 a t an ionic strength of 1.0 maintained with both LiC104 and NaC104. A summary of their rate and activation parameters is included in Table 111, together with those from this study. Using our data (Figure 4), the value of K,, for the monocyanochromium(II1) ion is estimated to be 0.1 1. mol-l a t an ionic strength of 1.0 maintained with sodium perchlorate. The corresponding product &K,, ( i e . , kl) was calculated from our data to be 3.2 X l o p 4 1. mol-' sec-', which is in good agreement with Birk and Espenson's value of k l of 1. mol-' sec-' (Table 111). 3.1 X Birk and Espenson treated the acid-catalyzed pathway in terms of a linear first-order dependence on hydrogen ion concentration. Assuming that K,, is 0.1 1. mol-' at an ionic strength of 1.0, a maximum deviation from linearity of only 10yGwould be predicted a t the highest hydrogen ion concentration used in their studies. It is not surprising, therefore, that such a small deviation from linearity was not observed or considered significant. 2 4 Furthermore, Deutsch and Taubelo showed that a plot of kobsd V S . [ H + ] does not vary significantly from linearity in the acid-assisted aquation of the acetatopentaaquochromium(II1) ion a t an ionic strength of 1.0 (lithium perchlorate), whereas a very significant deviation is observed at an ionic strength of 4.0. A t this ionic strength, they were able to calculate concordant values of the protonation equilibrium constant from both kinetic and spectral data. For the acid-catalyzed pathway, significant differences exist between the rate and activation parameters determined in the NaC104-HC104 medium and those in LiC104-HC104 solution (Table 111). I n both studies, the assumption that the activity coefficients of the reacting species remain constant as perchloric acid replaces the perchlorate salt is not (24) Separate experiments presently in progress in this laboratory by C. DeVine indicate t h a t a significant deviation from linearity is also observed in a plot of kobsd vs. [H +] for t h e monocyano complex a t I = 2.0 maintained with lithium perchlorate, T h e value of K,, appears t o be about t h e same a s t h a t in sodium perchlorate a t t h e same ionic strength.

AS*, eu

Ref

+12.7 ='c 4 . 2 -16.4 2.2 $ 5 . 4 f 3.1 -8.3 f 1 . 5 f 8 . 9 & 1.1 - 5 . 5 f0 . 3

5 5 5

*

5 This study

-16.0 f 4.0

rigorously correct and would be expected to hold true only when the proportion of the salt replaced by the acid is small. Biedermann and SillCn25have shown that the activity coefficients of a number of cations do remain constant in a medium of ionic strength 3.0 as sodium perchlorate is replaced by perchloric acid up to an acid concentration of 0.6 M. T o test our data for possible medium effects as the salt is replaced by acid, the rate parameters were calculated using the data from only the five lowest acidity points, then from the six lowest, and so on, until all of the acidity points were included. (For the data at 40°, this would mean that only those points below 0.4 M H + would be included in the first calculation.) At a given temperature the values of ko, k H , and K,, were found to vary randomly as more points a t higher acidities were included in the calculations and the values of the product kHK,, remained essentially constant throughout the range of acidities, varying less than 2%. We conclude, therefore, that there is no serious lack of constancy in the activity coefficients under the experimental conditions used ( I = 2.0) and that the curvature in the plot of kobsd V.Y. [ H + ]is actually the result of a protonation of the complex which increases with increasing ionic strength. On the basis of kinetic studies, monosubstituted chromium(II1) Complexes, Cr(H20)5X2+, can be divided into two groups, one of which includes complexes with a basic ligand such as cyanide, fluoride,7 azide,6 hypophosphitej8 sulfatejgand acetate,l0 while the other includes complexes with a nonbasic ligand such as chloride,' i ~ d i d e , ~ , isothiocyanate, '~,'~ l 1 nitrate, l 6 and per~ h l o r a t e . ' ~The complexes in the former group exhibit an acid-catalyzed aquation pathway, whereas those in the latter group are not acid catalyzed. Except for the cyanide complex, the complexes with a basic ligand exhibit an acid-catalyzed pathway with a linear first-order dependence on [H+]. For the cyanide complex the rate for this pathway is equal to { kxKeq * [H+]/(l K,,[H+]))Ct which reduces to k ~ K e q [ H + ] * Ct when Keq[H+]